A dual-task segmentation network based on multi-head hierarchical attention for 3D plant point cloud
Na minha lista:
| Publicado no: | Frontiers in Plant Science vol. 16 (Jul 2025), p. 1610443-1610459 |
|---|---|
| Autor principal: | |
| Outros Autores: | , , , , , , , |
| Publicado em: |
Frontiers Media SA
|
| Assuntos: | |
| Acesso em linha: | Citation/Abstract Full Text Full Text - PDF |
| Tags: |
Sem tags, seja o primeiro a adicionar uma tag!
|
MARC
| LEADER | 00000nab a2200000uu 4500 | ||
|---|---|---|---|
| 001 | 3273795140 | ||
| 003 | UK-CbPIL | ||
| 022 | |a 1664-462X | ||
| 024 | 7 | |a 10.3389/fpls.2025.1610443 |2 doi | |
| 035 | |a 3273795140 | ||
| 045 | 2 | |b d20250701 |b d20250731 | |
| 100 | 1 | |a Pan, Dan |u School of Electronics and Information, Guangdong Polytechnic Normal University, Guangzhou, China | |
| 245 | 1 | |a A dual-task segmentation network based on multi-head hierarchical attention for 3D plant point cloud | |
| 260 | |b Frontiers Media SA |c Jul 2025 | ||
| 513 | |a Journal Article | ||
| 520 | 3 | |a IntroductionThe development of automated high-throughput plant phenotyping systems with non-destructive characteristics fundamentally relies on achieving accurate segmentation of botanical structures at both semantic and instance levels. However, most existing approaches rely heavily on empirically determined threshold parameters and rarely integrate semantic and instance segmentation within a unified framework.MethodsTo address these limitations, this study introduces a methodology leveraging 2D image data of real plants, i.e., Caladium bicolor, captured using a custom-designed plant cultivation platform. A high-quality 3D point cloud dataset was generated through reconstruction. Building on this foundation, we propose a streamlined Dual-Task Segmentation Network (DSN) incorporating a multi-head hierarchical attention mechanism to achieve superior segmentation performance. Also, the dual-task framework employs Multi-Value Conditional Random Field (MV-CRF) to enable semantic segmentation of stem-leaf and individual leaf identification through the DSN architecture when processing manually-annotated 3D point cloud data. The network features a dual-branch architecture: one branch predicts the semantic class of each point, while the other embeds points into a high-dimensional vector space for instance clustering. Multi-task joint optimization is facilitated through the MV-CRF model.Results and discussionBenchmark evaluations validate the novel framework’s segmentation efficacy, yielding 99.16% macro-averaged precision, 95.73% class-wise recognition rate, and an average Intersection over Union of 93.64%, while comparative analyses confirm its superiority over nine benchmark architectures in 3D point cloud analytics. For instance segmentation, the model achieved leading metrics of 87.94%, 72.36%, and 71.61%, respectively. Furthermore, ablation studies validated the effectiveness of the network’s design and substantiated the rationale behind each architectural choice. | |
| 653 | |a Accuracy | ||
| 653 | |a Comparative analysis | ||
| 653 | |a Datasets | ||
| 653 | |a Deep learning | ||
| 653 | |a Conditional random fields | ||
| 653 | |a Ablation | ||
| 653 | |a Leaves | ||
| 653 | |a Semantic segmentation | ||
| 653 | |a Registration | ||
| 653 | |a Automation | ||
| 653 | |a Plants (botany) | ||
| 653 | |a Genotype & phenotype | ||
| 653 | |a Benchmarks | ||
| 653 | |a Efficiency | ||
| 653 | |a Vector spaces | ||
| 653 | |a Image reconstruction | ||
| 653 | |a Image segmentation | ||
| 653 | |a Computer vision | ||
| 653 | |a Phenotyping | ||
| 653 | |a Clustering | ||
| 653 | |a Three dimensional models | ||
| 653 | |a Effectiveness | ||
| 653 | |a Instance segmentation | ||
| 653 | |a Plant growth | ||
| 653 | |a Annotations | ||
| 653 | |a Morphology | ||
| 653 | |a Semantics | ||
| 653 | |a Environmental | ||
| 700 | 1 | |a Liu, Baijing |u School of Information Engineering, Guangdong University of Technology, Guangzhou, China | |
| 700 | 1 | |a Luo, Lin |u School of Computer Science and Technology, Guangdong University of Technology, Guangzhou, China | |
| 700 | 1 | |a Zeng, An |u School of Computer Science and Technology, Guangdong University of Technology, Guangzhou, China | |
| 700 | 1 | |a Zhou, Yuting |u School of Electronics and Information, Guangdong Polytechnic Normal University, Guangzhou, China | |
| 700 | 1 | |a Pan, Kaixin |u School of Electronics and Information, Guangdong Polytechnic Normal University, Guangzhou, China | |
| 700 | 1 | |a Xian, Zhiheng |u Guangzhou Huitong Agricultural Technology Co., Ltd., Guangzhou, China | |
| 700 | 1 | |a Yulun Xian |u Guangzhou Huitong Agricultural Technology Co., Ltd., Guangzhou, China, Guangzhou iGrowLite Agricultural Technology Co., Ltd., Guangzhou, China | |
| 700 | 1 | |a Liu, Licheng |u School of Information Engineering, Guangdong University of Technology, Guangzhou, China | |
| 773 | 0 | |t Frontiers in Plant Science |g vol. 16 (Jul 2025), p. 1610443-1610459 | |
| 786 | 0 | |d ProQuest |t Agriculture Science Database | |
| 856 | 4 | 1 | |3 Citation/Abstract |u https://www.proquest.com/docview/3273795140/abstract/embedded/H09TXR3UUZB2ISDL?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text |u https://www.proquest.com/docview/3273795140/fulltext/embedded/H09TXR3UUZB2ISDL?source=fedsrch |
| 856 | 4 | 0 | |3 Full Text - PDF |u https://www.proquest.com/docview/3273795140/fulltextPDF/embedded/H09TXR3UUZB2ISDL?source=fedsrch |